New Technologies Feed the Flames of Fake News
A 2017 Radiolab podcast investigates two new technologies currently in development that will allow for the creation of high quality and completely fake audio recordings and videos. An application currently being developed by Adobe will allow users to create an audio file from text using anyone’s voice — as long as you can get 40 minutes of audio recording of that person speaking to draw from.
Similarly, technologies under development will allow users to control the facial movements of a person on screen. By combining these two technologies, videos could be created of any person saying anything.
The technology isn’t quite ready to fool us yet — check out this video that Radiolab produced to test these techniques out: http://futureoffakenews.com/.
But, if this is what the Radiolab folks could whip up with minimal effort, how long will it be before we are fooled by these videos?
In a world of massive news illiteracy, how will visual and aural deception change the news landscape? One thing we seemed to have learned from the presidential election is that it doesn’t matter if media is determined to be fake at some point — if it reaches a lot of people, it does damage. So, even if news sources hire technologists who can weed out the deceptive videos, the existence of the deceptive videos will have an impact.
Beyond the fake news implications of these technologies, I got to thinking about more personal impacts. How might this affect the types of conversations I have over the phone? The amount of information I’m willing to disclose after recognizing a friend’s voice on the other end of the line? Might our technical abilities at some point cause us to revert back to pre-industrial forms of communication? To face-to-face conversations?
You can find the Radiolab episode “Breaking News” here: http://www.radiolab.org/story/breaking-news/